share-research-api

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements searching for research outputs via the official SHARE API provided by the Center for Open Science.
  • [EXTERNAL_DOWNLOADS]: The skill fetches research metadata from https://share.osf.io/api/v2. This is a recognized service for scholarly tracking and research aggregation.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted metadata from external research repositories.
  • Ingestion points: The Python script in SKILL.md reads and parses JSON data from the SHARE search endpoint.
  • Boundary markers: None explicitly provided to the LLM context.
  • Capability inventory: The skill is limited to literature search and does not possess dangerous capabilities such as file system writes, network exfiltration of local data, or execution of arbitrary commands.
  • Sanitization: The implementation uses specific JSON field extraction and truncates description strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 04:35 AM